University Of Alberta Fibre Type Specific Nuclear Localisation Patterns of NFAT- isoforms in CLFS-Induced Skeletal Muscle Adaptations By Pamela Caitlin McDonald A thesis submitted to the faculty of Graduate Studies and Research in partial fulfillment of the requirement for the degree of Master of Science Faculty of Physical Education and Recreation Pamela Caitlin McDonald Spring 2014 Edmonton, Alberta Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.
112
Embed
Fibre Type Specific Nuclear Localisation Patterns of NFAT ......LIST OF FIGURES CHAPTER ONE 1.1 The CLFS model and a summary of the CLFS-induced adaptations 17 1.2 A simplified diagram
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
University Of Alberta
Fibre Type Specific Nuclear Localisation Patterns of NFAT-
isoforms in CLFS-Induced Skeletal Muscle Adaptations
By
Pamela Caitlin McDonald
A thesis submitted to the faculty of Graduate Studies and Research
in partial fulfillment of the requirement for the degree of
Master of Science
Faculty of Physical Education and Recreation
Pamela Caitlin McDonald
Spring 2014
Edmonton, Alberta
Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is
converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms.
The author reserves all other publication and other rights in association with the copyright in the thesis and,
except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.
ABSTRACT
The purpose of this investigation was to characterize the nuclear
localisation patterns of NFAT-c1, -c2, -c3, and -c4 in relation to chronic low-
frequency stimulation induced fibre type transitions. A model has been
suggested to describe the roles of all 4 NFAT isoforms in activity-dependent
skeletal muscle fibre transformation. The findings of the present study are
consistent with different yet related roles for the NFAT isoforms in fast-to-
slow fibre type transformation. NFAT-c1 nuclear expression was associated
with transition toward slower fibre types (type-I, type-IIA, type-IID(X)),
whereas NFAT-c3 nuclear expression was associated with transformations
surrounding the type-IIA fibres. The nuclear expression patterns of NFAT-c2
and NFAT-c4 are consistent with their involvement in fibre maintenance and
cell survival.
ACKNOWLEDGEMENTS
I would like to thank first and foremost, my supervisor, Dr. Ted
Putman for his patience, expert guidance and overall support. In addition, I
would like to express my sincere gratitude to Karen Martins who first piqued
my interest in this area and has lent her expertise on more than one occasion.
Thank you to Ian MacLean and Karen Martins for all of their technical
assistance, without which this document would not be complete.
Furthermore, thank you to Stephen Ogg and Arlene Oatway for all of their
guidance in microscopy. Lastly, I would like to thank my committee Dr.
Walter Dixon and Dr. Darren Delorey for their time and direction throughout.
This work was funded by the Natural Sciences and Engineering Council of
Canada, Alberta Innovates – Health Solutions, and the University of Alberta.
Pette, 1993), and tropomyosin (Schachat, et al., 1988).
CLFS is known to induce changes in Ca2+ kinetics, as illustrated in
Figure 1.1. Firstly, the uptake characteristics of the sarcoplasmic reticulum
adapt to stimulation. Secondly, the amount of Ca2+-binding proteins in the
cytosol changes with CLFS, including the upregulation of phospholamban, a
protein of the sarcoplasmic recticulum Ca2+ -ATPase (SERCA), and the slow
SERCA2a isoform (Hu, et al., 1995; Ohlendieck, et al., 1991). There is a
concurrent down regulation of the fast SERCA1 isoform and Ca2+ binding
protein parvalbumin (Klug, et al., 1988). In addition, the ryanodine receptor,
dihydropyridine receptor, and triadin exhibit considerable reductions in
muscles stimulated with CLFS (Ljubicic, et al., 2005). Such adaptations lead
to the production of low amplitude Ca2+ transients (Carroll, S. L., et al., 1999),
in comparison to the phasic high amplitude Ca2+ transients in fast muscles,
that mimic those of endurance exercise and consequently cause activation of
slow fibre specific pathways (Figure 1.1). The slow biphasic pattern of Ca2+
transients is known to activate certain downstream signalling pathways
15
including the Cn- NFAT pathway (Chin, et al., 1998; Martins, et al., 2012; Wu,
et al., 2000).
Additionally, CLFS induces changes in the metabolic pathways of ATP
production that are characterised by increased expression and activity of
mitochondrial enzymes that catalyse terminal glucose and fat oxidation
(Figure 1.1)(Putman, et al., 2004); this adaptive change is a direct function of
mitochondrial genesis (Ljubicic, et al., 2005). CLFS also induces a concurrent
decrease in the expression and activity of glycolytic enzymes (Hood, et al.,
1989; Pette & Vrbova, 1999). The change in oxidative enzyme activity in
response to CLFS has been inversely correlated with basal levels of a muscle
(Ljubicic, et al., 2005). In other words, muscles that already report high
levels of oxidative activity at rest do not show as great of an increase as
highly glycolytic muscles.
An increase in capillary density (Putman, et al., 2001; Skorjanc, et al.,
1998) and myoglobin content (Kaufmann, et al., 1989) occur quite rapidly in
response to CLFS, providing vascular support for the new, smaller, more
oxidative, fibres as illustrated in Figure 1.1. Slower fibres (e.g., type-I), in
general, have a smaller cross sectional area than the larger fast fibres (e.g.,
type-IIB). For the muscle to be able to support the smaller cross sectional
area of the new slower fibres, muscle satellite cells are known to actively
cycle and form the basis for creation of new myonuclei (Martins, et al., 2006;
Martins, et al., 2009; Putman, et al., 1999, 2000). As well, achieving a
threshold myonuclear domain seems to be required to allow for fibre type
transition to proceed (Pavlath, et al., 1989). The combination of a smaller
cross sectional area and an increased capillary density, seem to be required
to support the new slower-phenotype (Brown, et al., 1976; Pette, et al., 1975;
Skorjanc, et al.).
16
Therefore, by using CLFS as a model of muscle training the inherent
plasticity of skeletal muscle can be readily observed. The chronic nature of
CLFS provides a predictable environment that stimulates metabolic,
structural, enzymatic, and conformational changes of the muscle.
Encompassing all aspects involved in muscle structure and function. The use
of CLFS to study the cellular and molecular mechanisms that underlie
skeletal muscle plasticity removes the two major limitations of exercise
training models. First, muscle damage is not a factor with CLFS in rat.
Second, motor units are recruited maximally and synchronously, which
involves inducing muscle changes in all fibres.
17
Figure 1.1 The CLFS model and a summary of the CLFS-induced adaptations. Ach=acetylcholine; Ca2+=calcium; CLFS=chronic low-frequency stimulation; CS=citrate synthase; COX=cytochrome c oxidase; EDL=extensor digitorum longus; GAPDH=glyceraldehyde phosphate dehydrogenase; NMJ=neuromuscular junction; PFK= phosphofructokinase; TA=tibialis anterior. With permission from Ljubicic et al. 2005 .
The underlying signalling mechanisms that link slow tonic motor nerve
activity to shifts in fibre type specific gene expression (i.e., MyHC), have yet to
be fully elucidated. Slow motoneuron activity is known to trigger sustained,
low-amplitude increases in intracellular Ca2+ transients, which not only
result in contractile activity but also stimulate a number of transcriptional
factors and co-activators. The two major pathways that are known to be
activated by slow motoneuron activity are Myocyte Enhancer Factor (MEF)2-
Class II histone deacetylase proteins (HDAC), and the focus of this
investigation, the Cn-NFAT pathway. It has been postulated that these
pathways interact in the cell to activate slow specific genes in response to
low-frequency stimulation.
Ca2+ concentration inside the cell is controlled largely by the
sarcoplasmic reticulum through calcium release and uptake systems in
skeletal muscle. Ca2+ transient kinetics have a direct effect on fibre
properties, indicated by the higher level of cytosolic free Ca2+ concentration
in slow muscles (Carroll, S. L., et al., 1997), as well as the lower peak and
slower decline of Ca2+ transients in slow muscles. One explanation for the
contrast in [Ca2+] between fibre types is a more developed, larger
sarcoplasmic reticulum terminal cisternae in fast muscles leading to a higher
amplitude Ca2+ influx with excitation. One similarity between fibre types is
the rate of decline of amount of Ca2+ released in subsequent action potentials.
Overall, Ca2+ kinetics differ between fast and slow muscles as well as with
different types of stimuli (i.e., endurance vs. resistance training), in turn
activating different pathways and promoting different gene programs. (For a
recent review see Schiaffino & Reggiani, 2011).
19
With chronic stimulation of skeletal muscle there is a sustained influx
of intracellular Ca2+, which in turn causes the phosphorylation of Class II
HDAC proteins by nuclear calmodulin dependent kinase (CaMK).
Subsequently, HDAC binds to 14-3-3 protein and moves out of the nucleus
into the cytoplasm (McKinsey, et al., 2001), this removes the inhibition of
MEF2 and allows transcriptional activity to occur. Two additional proteins
have been shown to cause slow gene upregulation as well as the activation of
MEF2, AMPK (activated protein kinase) and Calcineurin (Cn) (as shown by
MEF2 blockage caused cyclosporine A (Wu, et al., 2000)). The specific
downstream pathways of MEF2 are still largely unknown. However, it has
been elucidated that MEF2 binds to an A/T-rich DNA sequence that is in the
control regions of numerous muscle specific genes, particularly in the same
regions of slow fibre specific genes (similar to NFAT). (For recent review see
Bassel-Duby & Olson, 2006).
The same chronic influx of Ca2+ that activates HDAC also stimulates
calmodulin and the B subunit of Cn, which in turn dephosphorylates nuclear
factor of activated T-cells (NFAT), and subsequently translocates into the
nucleus (McCullagh, et al., 2004). Once in the nucleus, NFAT binds to specific
sequences on the promoter regions of specific target genes that induce a
slower fibre-specific gene program (Liu, Y., et al., 2005)(Figure 1.2). The
activity-induced up-regulation of slow fibre specific genes is dependent on
Cn-activation and NFAT-c1 nuclear import. This pathway also demonstrates
properties that implicate Ca2+ signalling in mediating the adaptive response.
Carroll et al. (1999) reported that transforming single fibres exposed to CLFS
displayed [Ca2+]i transients typical of slow-twitch fibres, with characteristic
biphasic low-amplitude contraction-dependent changes in intracellular Ca2+
concentration ([Ca2+]intracellular) on the order of 1.2-1.5 µM. This
[Ca2+]intracellular coincides with the concentration that is required for half-
maximal activation of Cn-NFAT signalling pathway by [Ca2+]intracellular (Healy,
et al., 1997). Indeed a previous study from our laboratory (Martins, et al.,
20
2012) confirmed that activity-induced up-regulation of MyHC-I expression
was dependent on Cn-activation and NFAT-c1 nuclear import (As seen in
Figure 1.2).
Figure 1.2 A simplified diagram of the Ca2+/calcineurin/NFAT pathway for activation of slow skeletal muscle fibre gene expression. An action potential in the external membranes causes elevated cytosolic Ca2+, which then activated Cn, which subsequently dephosphorylates cytoplasmic NFAT. NFAT is then permitted to enter the nucleus, leading to the activation of slow fibre genes by nuclear NFAT.
21
1.4.1 Calcineurin
Calcineurin (Cn), a cyclosporine-sensitive, Ca2+-regulated serine/threonine
phosphatase, is a heterodimer of a calmodulin-binding catalytic subunit
(CnA) (59 to 62 kDa) and a Ca2+-binding regulatory subunit (CnB) (19 kDa),
each with various isoforms that are transcribed from different genes
(Parsons, et al., 2003). Three catalytic genes (CnA , CnA & CnA ) have been
discovered, however only two are expressed in skeletal muscle (CnA and
CnA ) (Parsons, et al., 2003). Whilst two regulatory subunits (CnB1 & CnB2)
have been discovered, only one is expressed in skeletal muscle (CnB1)
(Parsons, et al., 2004).
In response to a sustained low-amplitude increase in intracellular Ca2+
transients, as seen in response to sub-maximal exercise and CLFS, Ca2+ binds
with the CnB subunit and assists in the sub-maximal activation of the CnA
subunit. The same Ca2+ transients reportedly free calmodulin to bind Ca2+ at
high affinity sites, and subsequently to an auto-inhibitory region near the C-
terminus of Cn-A , thereby fully activating the enzyme complex. Once
activated, the Cn-enzyme complex proceeds to bind to and dephosphorylate
multiple phospho-serine and phospho-threonine residues on local NFAT-
isoforms, thereby facilitating their translocation into the nucleus (Figure 1.3)
(Al-Shanti & Stewart, 2009). Once in the nuclear compartment an NFAT
isoform may cooperatively bind to promoter regions with other transcription
that determine slow or fast muscle fibre phenotypes. Evidence to date would
seem to indicate that within both the Cn-enzyme complex and its target
NFAT-isoforms there is considerable potential for graded responses to
activity-dependent variations in [Ca2+]i that might account for precise
adjustments in Cn-NFAT signalling and control MyHC-based muscle fibre
types, especially during CLFS-induced fast-to-slow fibre type transformations.
22
Calsarcins (Frey, et al., 2008; Frey, et al., 2000) are a family of
sarcomeric proteins that tether Cn to the -actinin at the z-line of the
contractile apparatus, and regulate Cn activity. There are three calsarcin
isoforms, (calsarcin-1, calsarcin-2 and calsarcin-3) each having slightly
different roles in skeletal muscle. Calsarcin-2, and calsarcin-3 are specifically
expressed in fast-twitch muscle, whereas Calsarcin-1 is only expressed in
cardiac and slow skeletal muscle. Calsarcin-2 knockout caused an increase in
slower phenotype muscles as well as increased running distance in mice
(Frey, et al., 2008). Thus, Calsarcins seem to position Cn in a unique position
near a large intracellular Ca2+ pool to foster their activation with many up
and downstream factors, whilst hindering Calcineurins activity by tethering
it to the z-line, rendering it unable to translocate.
23
Figure 1.3 The structure of calcineurin and its enzymatic activation.
(A) Calcineurin consists of two subunits: catalytic-regulatory subunit (calcineurin A) containing the N-terminal catalytic and C-terminal regulatory domains. The C-terminal regulatory domain consists of three domains, which are the calcineurin binding site, calmodulin binding site and an auto-inhibitory site; and the Ca2+-binding regulatory subunit (calcineurin B).
(B) In resting skeletal muscle cells the intracellular Ca2+ level is low, and therefore the heterodimeric structure of calcineurin is in an inactive conformational state. In response to physical activity and exercise intracellular Ca2+ levels increase and activate calmodulin, which, in turn, binds to the heterodimer calcineurin, inducing its active conformational changes.
(C) In its active stable conformational state the autoinhibitory domain has been displaced from the catalytic domain where the nuclear factor of activated T-cell (NFAT) protein binds. The binding of Ca2+ to the binding regulatory subunit, calcineurin B, stabilizes active Cn. With permission from (Al-Shanti & Stewart, 2009).
24
1.4.2 Review of Nuclear Factor of Activated T-cells (NFAT)
1.4.1.1 STRUCTURE
NFAT was first discovered in T-cells as a rapidly inducible transcription
factor regulated by Ca2+-activated signalling pathways (Shaw, et al., 1988).
Historically related to REL-nuclear factor- B (REL-NF- B) family of
transcription factors, NFAT has since been shown to have a contrary role in
T-cells. The Ca2+-regulated isoforms of NFAT seem to play a role in T-cell
apoptosis (Lee, H., et al., 2005), whilst NF- B is active in cell reproduction.
There are four isoforms of calcium activated NFAT: NFAT-c1 (also known as
NFAT2 or NFATc); NFAT-c2 (also known as NFAT1 or NFATp); NFAT-c3
(also known as NFAT4 or NFATx); NFAT-c4 (also known as NFAT3). There is
a fifth isoform (NFAT5) that has a structure 40% similar to the Ca2+-activated
NFAT isoforms (Lopez-Rodriguez, et al., 1999), but is activated in response to
osmotic stress and has not been found active in the pathways being
investigated. Throughout, NFAT will refer to the four calcium activated
NFAT isoforms.
Structurally the four NFAT isoforms are quite alike (Figure 1.4). They
contain a high degree of similarity among the RHR (Rel-homology region),
which includes the regulatory region (Luo, et al., 1996) and the DNA binding
region (Jain, et al., 1995), suggesting common DNA binding specificities and
partner interactions. Many of the RHR-C interface residues observed in
NFAT-c2 are conserved in NFAT-c1 and NFAT-c3. NFAT-c4, however, seems
to have different residues in many of those positions (Hogan, et al., 2003).
The similarities between NFAT-c1, -c2, and -c3 might explain the redundancy
in some NFAT related functions. It has been suggested that regulating the
balance between specific NFAT isoforms occupying a specific promoter at
any given time may be controlling some function. The N-terminal and C-
terminal transactivation domains, on the other hand, vary between isoform.
There has been an abundance of research showing distinct roles for each
25
NFAT isoform in many different cell types (For a review see Rao et al., 1997)
leading to the idea that each isoform would have different roles in skeletal
muscle. Within the regulatory region of NFAT there are 13 activation sites
(Okamura, et al., 2000). Due to the number of phosphorylation sites,
multisite phosphorylation (Hogan, et al., 2003) could increase the sensitivity
of each NFAT isoform to stimuli through the varying thresholds of activation
of the many serine residues on the different NFAT isoforms. This variance in
activation sites as well as the differences in C-terminus and N-terminus
regions could explain the difference in response to stimulation and Ca2+
influx.
Figure 1.4 The General structure of nuclear factor of activated T cells (NFAT) transcription factors. NFAT proteins consist of an amino-terminal regulatory domain (also known as an NFAT homology region (NHR)), a DNA-binding domain (also known as a REL-homology domain (RHD)) and a carboxy-terminal domain. The regulatory domain contains an N-terminal transactivation domain (TAD), as well as a docking site for for casein kinase 1 (CK1), termed FSILF, and for calcineurin, termed SPRIEIT. It also includes multiple serine-rich motifs (SRR1, SP1, SP2, SRR2, SP3 and KTS) and a nuclear localisation sequence (NLS). With permission from (Muller & Rao, 2010).
26
1.4.1.2 FUNCTION
The four calcium activated NFAT isoforms exhibit varying actions throughout
the body, including cell differentiation, growth promotion, cell apoptosis and
cell production. All NFAT isoforms are dephosphorylated by a Ca2+-Cn
dependent pathway, and at least one isoform is active in most cell types.
Specifically, NFAT-c2 controls bone resorption by stimulating the
differentiation and functioning of osteoclasts (Ikeda, et al., 2006). In
melanoma cells, NFAT-c2 has an anti-apoptotic effect (Perotti, et al., 2012).
Whereas, NFAT-c2 deletion reduces cytokine production by mast cells
(Solymar, et al., 2002; Tsytsykova & Goldfeld, 2000). In adult rat joints,
NFAT-c2 acts as a negative regulator to promote appropriate cartilage
formation (Horsley and Pavlath 2002). The rate limiting step of NFAT-c2
with Ca2+ activation is its nuclear export, slow deactivation causes sustained
responses after termination of initial signal (Kar, et al., 2011). Overall, NFAT-
c2 seems to have functions ranging from pro- to anti-apoptotic, however in
general seems to be involved in cell regulation.
Similar to NFAT-c2, NFAT-c1 seems to have a predominant role in the
differentiation of osteoclasts from monocyte precursors (Takayanagi, et al.,
2002) and the differentiation of Th2 cells from antigen- “naive” T-cells. (for
review see Glimcher & Murphy, 2000). NFAT-c1 is the only isoform that is
regulated by a positive auto-regulation loop (Zhou, et al., 2002), and as such
could explain why it is the most commonly identified regulator in many cell
types. Deletion of NFAT-c1 causes defects in cardiac valve formation and
lethality (de la Pompa, et al., 1998; Ranger, et al., 1998). NFAT-c1 is unable to
promote apoptosis in T-cells by itself (Chuvpilo, et al., 2002). NFAT-c1’s
transcriptional activity could, in turn, be dependent on the other isoforms
activity in addition to its own.
27
Disruption of NFAT-c4 causes no known phenotype to develop in
vascular tissue (Graef, et al., 2001). In other tissues NFAT-c4 has been shown
to be involved in cell survival (Vashishta, et al., 2009) which could be the case
in skeletal muscle as well. In other tissue types, NFAT-c4 has been
implicated in cell apoptosis (Li, L., et al., 2013), and may underlie activity-
dependent neuronal plasticity throughout the adult brain (Bradley, et al.,
2005). In regards to kinase activity, the knockdown of GSK3 significantly
increased depolarization induced nuclear translocation of NFAT-c4 in
neurons, while inhibition of mTOR and p38 had no effect and nuclear
localisation only occurred after prolonged Ca2+ increases. When NFAT-c4
protein is ubiquitinated, decreases in protein levels and transcriptional
activity occur. Additionally, GSK3 enhanced ubiquitination and decreased
transactivation in cardiac tissue (Fan, et al., 2008), revealing a possible
mechanism by which NFAT-c4 is controlled.
NFAT-c3 is rapidly dephosphorylated and translocated into the
nucleus in response to Ca2+ in neuronal cells (Ulrich, et al., 2012), which
seems to be the case in most cell types. In response to induced hypoxia,
which causes an increase in arterial reactive oxygen species, NFAT-c3
activates and there is increase in Ca2+ and vasoconstrictor reactivity
(Friedman, et al., 2013). High extracellular Ca2+ induces activation and
expression of NFAT-c1, which in turn increases NFAT-c3 expression and
activity in osteoblasts (Lee, H. L., et al., 2011). There are some additional
instances where two or more NFAT isoforms work together or have
redundant roles. For instance, deletion of both NFAT-c1 and NFAT-c2 causes
loss of cytokine production in T-cells (Peng, et al., 2001). Also, in T-cells
NFAT-c2 and NFAT-c3 have very similar responses to Ca2+ influx and
inhibition. (Rao, et al., 1997).
In resting skeletal muscle, inactive NFAT is heavily phosphorylated
and, for the most part, located in the cytoplasm. NFAT is dephosphorylated
28
by the Cn-enzyme complex, which binds to the conserved regulatory domain,
followed by translocation into the nucleus where it co-ordinately binds with
other transcription factors, and to specific promoter regions that up-regulate
skeletal muscle specific genes (Meissner, Umeda, et al., 2007). NFAT protein
activation closely follows the activation pattern of Cn. NFAT-c1 is the most
studied isoform in skeletal muscle to date. It has been found to be involved
in the transition to a slower phenotype, especially to the slowest (ie., type-I)
fibres. Research has been completed on the varying roles of NFAT in the
growth and development of skeletal muscle. Knockout mice have been a
common model used to study each NFAT isoform separately. NFAT-c2
knockout has been shown to effect fibre cross sectional area during muscle
growth, as well as myonuclear number (Horsley, et al., 2001). NFAT-c3
knockout, on the other hand, reduced number of fibres in developing muscle
(Kegley, et al., 2001). Other knockout models have shown to be lethal and
therefore not possible to test in this manner.
Recently, using sodium bicarbonate buffering and heat in cultured
human muscle cells, Yamaguchi, et al. (2013) found contradictory results to
many of the previous studies. While, bicarbonate buffering and increased
heat have been found to cause an increase in slower muscle fibres,
Yamaguchi, et al. (2013) found an increase in NFAT-c2 and NFAT-c4
concurrent with an increase in the slower fibre types under these conditions.
Differences between this study and others could be due to the model used;
some studies have shown the coordinate increase in calcium with the
decrease of proteins that shuttle NFAT out of the nucleus is needed to
discover a substantial increase. Other pathways could be the major regulator
to adaptations with heat and bicarbonate buffering.
NFAT is rephosphorylated by several protein kinases such as glycogen
synthase kinase-3 (GSK-3 )(Beals, Sheridan, et al., 1997), protein kinase A
kinase (JNK) (Chow, et al., 2000), which promote its nuclear exit (Shen, et al.,
2006). Some act as export kinases, some as maintenance and some as both.
CK1, for example, docks at a conserved motif near the N terminus and acts as
a maintenance and export kinase for SRR1. GSK3 , on the other hand,
functions as an export kinase for the SP1 motif of NFAT-c2, and the SP2 and
SP3 motifs of NFAT-c1 (Beals, Clipstone, et al., 1997). It has been suggested
that with NFAT-c1, substrate sites for GSK3 might only be created after
previous phosphorylation of SP1 by cyclic-AMP-dependent protein kinase
(PKA) (Sheridan, et al., 2002) or dual-specificity tyrosine-phosphorylation-
regulated kinase (DYRK)-2 (Muller, et al., 2009). In addition, Drenning et al.
(2009) illustrated that calcium-dependent Nitric Oxide (NO) inhibits GSK-3
in a cGMP/PI3K/Akt dependent manner, and in turn assists in NFAT-c1’s
function. Martins et al. (2012) corroborated these findings in showing that a
reduction in NFAT-c1 activity is caused by obstructing GSK-3
dephosphorylation through the inhibition of NO.
Variations in resting nuclear influx and efflux have been found
between the different NFAT isoforms. NFAT-c1, for example, has been
shown to shuttle in and out of nucleus under resting conditions, and
accumulate in the nucleus with the blockage of chromosome region
maintenance 1 (CRM1) (Shen, et al., 2006). NFAT-c1 nuclear import and exit
occurred even in the presence of other kinase inhibitors, and the blockage of
Cn (Shen, et al., 2006). Shen et al. (2006) suggests that resting shuttling of
NFAT-c1 is constantly occurring regardless of phosphorylation status, whilst
NFAT-c3 does not accumulate in the nucleus even with the blockage of CRM1
(Shen, et al., 2006). Furthermore, it could be suggested that either a different
pathway regulates nuclear efflux, or simply there is limited resting shuttling
of NFAT-c3. MAPKs have been postulated to differentially phosphorylate the
first serine of SRR1 depending on NFAT isoform. For example, p38 MAPK
has been show to selectively phosphorylate NFAT-c2 (Gomez del Arco, et al.,
30
2000), whilst JNK phosphorylates NFAT-c1 and NFAT-c3 (Chow, et al., 1997).
Selective action of the various NFAT phosphorylation kinases could be one
major factor in the different actions and functions demonstrated by the
different NFAT isoforms. Overall, many factors effect each NFAT isoform
leading to distinct downstream pathways in skeletal muscle. (For a review
see Hogan et al., 2003).
1.4.2 The Calcineurin-NFAT Pathway
The Cn-NFAT pathway has been shown to play a key role in fibre type
regulation, particularly fast-to-slow transformation. Chin et al. (1998)
provided initial evidence through the use of skeletal myocytes that were
transfected with a constitutively active (*) form of Cn*. The activation of Cn
resulted in the selective upregulation of slow fibre specific gene promoters,
which were mediated by NFAT. Conversely, the inhibition of Cn activity, by
exposure to the Cn-inhibitor cyclosporine A (CsA), caused the opposite slow-
to-fast fibre type transformation (Chin, et al., 1998). Numerous studies have
subsequently corroborated the involvement of the Cn-NFAT pathway in the
promotion of the slow fibre-specific gene program in vitro (Delling, et al.,
2000; Higginson, et al., 2002; Meissner, Umeda, et al., 2007; Naya, et al., 2000;
Torgan & Daniels, 2001) and in vivo (McCullagh, et al., 2004; Serrano, et al.,
2001; Talmadge, et al., 2004). For example, Meissner, et al. (2007)
demonstrated that exposure of differentiated C2C12 myotubes to a Ca2+-
ionophore or transfection with constitutively active NFAT-c1 induced
increases in MyHC-I promoter activity, whilst exposure of these same
myotubes to CsA blocked MYHC-I promoter activity. Similarly, studies by
Talmadge, et al. (2004) revealed that the diaphragm muscles of transgenic
mice that expressed high levels of constitutively active Cn (MCK-Cn* mice)
had elevated expression levels of MyHC-I and lower expression levels of both
MyHC-IIa and MyHC-IId(x). These collective results support a role for the
Cn-NFAT pathway in regulating of slow fibre gene programs.
31
The Cn-NFAT pathway has also been implicated in fast-to-slow fibre
transformation within the subpopulations of fast fibres (Dunn, et al., 1999).
For example, (Allen, David L., et al., 2001) observed that the over-expression
of Cn in C2C12 myotubes promoted MyHC-IIa expression; the over-
expression of NFAT-c also induced MyHC-IIa expression, but was further
associated with lower expression levels of MyHC-IId(x) and MyHC-IIb. Allen
& Leinwand (2002) later showed that exposure of C2C12 myotubes to a Ca2+-
ionophore activated the MyHC-IIa promoter to a much greater extent than
the promoters regulating MyHC-IId(x) or MyHC-IIb expression. Thus, Ca2+-
ionophore treated myotubes clearly showed a shift toward a slower
phenotype within fast muscle fibres in response to changes in intracellular
concentrations of Ca2+ that were typical of slow type stimulation. The results
of those studies also provide substantial evidence that Cn and NFAT-c are
downstream regulatory targets and critical regulatory steps in mediating
fast-to-slow fibre type transitions.
Dunn et al. (2000) uses a transgenic mouse model of Cn
overexpression (CnA*) as well as, cyclosporine A administration (CsA; which
blocks Cn) and overload of the mouse plantaris m. by synergist ablation to
study fast-to-slow fibre type transformation. The overloaded mice,
regardless if they were CnA* or wild type, were shown to have an increase in
fibres expressing MyHC-IIa and MyHC-I isoforms, as well as an increase in
slow fibre genes there was a concurrent decrease in fast fibre genes. Which
was confirmed by Rana et al (2008), who found a NFAT binding site on the
troponin I fast (TnIf) intronic regulatory element (FIRE). Conversely, with
CsA administration the Cn-NFAT pathway was blocked, evidenced by the
decrease in NFAT-c1 in this group, which causes a blockage of the fast-to-
slow conversion that regularly occurs. Together, the results of Dunn, et al.
(2000) support a role for the Cn-NFAT in fast-to-slow transformation within
the fast population, as well as the final fast-to-slow transformation (MyHC-IIa
to MyHC-I) in the plantaris m. Dunn, et al. performed similar experiments in
32
(1999) and (2001) which produced the same findings as the 2000 paper.
Altogether, the Ca2+-Cn-NFAT pathway is shown to contribute to the fast-to-
slow myosin heavy chain transformation that occurs in response to exercise
in the fast population. Most recently, Vechetti-Junior, et al. (2013) reported
an increase in NFAT-c1, -c2, and -c3 without a concurrent increase in Cn
activity with 8-weeks of aerobic training. One explanation for this could be
the fact that they are looking at Cn mRNA levels after 8-weeks of training,
and Cn has been shown to be a precursor to NFAT activity as well as fibre
type transition. That late of a time point with training the majority of the
adaptation should have already occurred.
A recent study reported that the four Ca2+-Cn dependent NFAT
isoforms (NFAT-c1, NFAT-c2, NFAT-c3, and NFAT-c4) are responsible for
activating different fibre type genes and act co-ordinately with each other
downstream of Cn (Figure 1.5) (Calabria, et al., 2009). Using a loss-of-
function approach by mRNA-inhibition (RNAi) in rats, Calabria et al. (2009)
found that all four NFAT isoforms were necessary for MyHC-I expression. In
contrast, the expression of MyHC-IIa and MyHC-IId(x) required only three
NFAT isoforms (i.e., NFAT-c2, -c3, and -c4). Notably there was a stronger
relative influence of NFAT-c3 on MyHC-IIa expression than on MyHC-IId(x)
expression. Finally, NFAT-c4 alone was required for MyHC-IIb expression.
Examination of rodent muscles at rest has revealed distinct expression
pattern of NFAT isoforms within muscle fibres. Fractionation experiments
(Calabria, et al., 2009) using muscles composed of different fibre types (i.e.,
soleus m. and extensor digitorum longus m.) provide evidence that NFAT-c4
resides in a predominately nuclear location in all fibre types. By comparison,
NFAT-c2 and -c3 display low but detectable nuclear signals in both muscles,
whilst NFAT-c1 is in a predominately nuclear position in the soleus m. but not
in the extensor digitorum longus m. A more detailed analysis by Mutungi
(2008) found that NFAT-c1 occupies a mainly cytoplasmic location in type-
IIA fibres in both mixed fast-twitch extensor digitorum longus m. and slow-
33
twitch soleus m. of rat. In the type-I fibres NFAT-c1 was found to primarily
reside within a nuclear position, again regardless whether the fibres were in
a mainly fast or slow muscle.
In addition, Calabria et al. (2009) applied stimulation to their samples,
the slow pattern of stimulation (200 pulses, 20 Hz every 30 seconds) caused
nuclear accumulation of NFAT-c1, -c2, and -c3, where NFAT-c4 was already
nuclear in location. The fast pattern of stimulation (25 pulses, 100 Hz every
60 seconds) caused nuclear accumulation of NFAT-c4, -c3 and in some cases -
c2. The patterns of NFAT nuclear accumulation that are caused by particular
patterns of stimulation are consistent with earlier findings in resting muscle
linking specific NFAT isoforms with certain MyHC isoforms (Calabria, et al.,
2009). The dephosphorylation and translocation of NFAT-c1 by a chronic
low-frequency pattern of stimulation has been demonstrated previously
(McCullagh, et al., 2004) but the translocation of the specific NFAT isoforms
in relation to fibre type has yet to be elucidated.
34
Figure 1.5 A graded nuclear recruitment of NFAT transcription factors in response to activity/exercise changes the expression of MyHC isoforms. c1=NFAT-c1, c2=NFAT-c2, c3=NFAT-c3, c4=NFAT-c4. A slow pattern of activity induces the nuclear translocation and activation of all 4 NFAT isoforms, which are necessary for the transcription of MyHC-I. NFAT-c1, in particular, only translocates to the nucleus in response to slow-type activity (Serrano, et al., 2001), (Naya, et al., 2000) and switches the specificity of the transcriptional machinery toward slow MyHC; by inhibiting fast MyHC expression (Chin, et al., 1998), it also ensures that the fast and slow gene programs are mutually exclusive. In fast-oxidative (IIA/IIX) fibres, NFAT-c1 is mostly cytosolic and NFAT-c2 and -c3 are transcriptionally active; thus, contributing to the transcriptional activation of MyHC-IIa and -IIx. In fast-glycolytic fibres, where MyHC-IIB is the only transcribed isoform, only NFAT-c4 is nuclear and transcriptionally active. With permission from Calabria et al., 2009.
35
1.5 Thesis Objectives
Using selective molecular RNAi knockdown approach Calabria, et al. (2009)
reported that fast-glycolytic type-IIB fibres are only transcriptionally
dependent on NFAT-c4, whilst expression of fast-oxidative-glycolytic (FOG)
type-IID/X, or fast-oxidative (FO) type-IIA fibre phenotypes was permissively
dependent on NFAT-c4 but transcriptionally dependent on both NFAT-c3 and
NFAT-c2. They further reported that slow-oxidative (SO) type-I fibres were
only transcriptionally dependent on NFAT-c1. However, physiological
studies conducted by this group could only demonstrate that myonuclear
export of NFAT-c1 was stimulated by fast motor-unit like direct nerve
stimulation (i.e., 100 Hz), whilst NFAT-c1 myonuclear import was stimulated
by slow motor-unit like direct nerve stimulation (i.e., 20 Hz). Indeed, other
studies have confirmed that NFAT-c1 nuclear translocation and
transcriptional activation are required for activity-induced fast-to-slow fibre
type conversions (Chin, et al., 1998; Martins, et al., 2012). These collective
findings suggest different roles for the various NFAT-isoforms in determining
muscle fibre types, and possibly in regulating activity-induced fast-to-slow
fibre type transformations that are dependent on Cn activation.
The present study tested the following hypotheses:
A. Within unstimulated rat peroneus longus m.: (1) type-I fibres will
uniquely display the greatest NFAT-c1 myonuclear localisation, but will
also express NFAT-c2, -c3, -c4 at levels similar to all other fibre types;
(2) type-IIA, and -IID/X fibres will display the greatest frequency of
NFAT-c2 and NFAT-c3 myonuclear localisation, but will also express
NFAT-c4 similar to type-IIB fibres. (3) type-IIB fibres will only display
NFAT-c4 nuclear localisation.
36
B. Application of CLFS to the peroneus longus m. will induce sequential
step-wise fast-to-slow fibre type transitions in the direction of
IIBIID/XIIAI and will correspond to the NFAT-isoform profile
unique to the slower fibre type at each transitional junction according to
the next-nearest neighbour rule. (1) CLFS-induced up-regulation of
MyHC-IIa and -IId(x) within transforming fibres will be preceded by
greater myonuclear localisation of NFAT-c2 and NFAT-c3. (2) CLFS-
induced up-regulation of MyHC-I within transforming fibres will be
preceded by greater myonuclear localisation of NFAT-c1.
37
CHAPTER TWO
MATERIALS AND METHODS
38
2.1 Materials
Standard lab reagents were purchased from Fisher Scientific (Ontario,
Canada), Sigma-Aldrich Canada (Oakville, ON, Canada), Vector Laboratories,
Inc. (Burlingame, CA, USA). All other materials are listed in the following:
Table 2.1 Primary Monoclonal Anti-Myosin Heavy Chain Antibodies
Harvested from Mouse Hybridoma Cells Obtained From the
American Type Culture Collection (ATCC)
Antibody Clone Number Antibody Specificity
BA-D5 IgG - anti-MyHC-I
SC-71 IgG - anti-MyHC-IIa
BF-F3 IgM - anti-MyHC-IIb
BF-45 IgG - anti-MyHC-embryonic
BF-35 IgG - all MyHC isoforms but not MyHC-IIx (IId)
Table 2.2 Monoclonal (mc) and Polyclonal (pc) Primary Antibodies
Used for Immunohistochemistry
Antibody Clone/Product Number – Class – Source
mc anti-dystrophin DYS2 (Dy8/6C5) – IgG1 – Novocastra Laboratories, Newcastle, UK
pc anti-laminin 68125 – IgG – ICN Biomedicals Inc., Aurora, OH, USA
pc anti-NFAT-c1 H-110, sc-13033 – IgG – Santa Cruz Biotechnology, CA, USA
pc anti-NFAT-c2 M-300, sc-13034 – IgG – Santa Cruz Biotechnology
pc anti-NFAT-c3 M-75, sc-8321 – IgG – Santa Cruz Biotechnology
pc anti-NFAT-c4 H-74, sc-13036 – IgG – Santa Cruz Biotechnology
39
Table 2.3 Affinity-Purified Biotinylated Secondary Antibodies Used
for Immunohistochemistry
Antibody Species
Host Clone/Product Number – Source
anti-mouse IgG (rat-absorbed)
horse BA-2001 – Vector Laboratories, Inc. (Burlingame, CA, USA)
anti-mouse IgM goat BA-2020 – Vector Laboratories, Inc. (Burlingame, CA, USA)
anti-rabbit IgG horse BA-1000 – Vector Laboratories, Inc. (Burlingame, CA, USA)
Table 2.4 Software
Name Source
Pette Program Dirk Pette (Konstanz, Germany)
Statistica 8 Statsoft Inc. (CA, USA)
Sigma Plot 12.5 Systat Software, Inc. (CA, USA)
Table 2.5 Miscellaneous
Name Source
Avidin-D Biotin Blocking Kit Vector Laboratories, Inc. (Burlingame, CA, USA).
Vectastain kit Vector Laboratories, Inc. (Burlingame, CA, USA).
40
2.2 Methods
2.2.1 Animal Care
The peroneus longus muscles used in the present study were previously
excised during two experiments (Martins, et al., 2006; Martins, et al., 2009).
Twenty-four Male Wistar Rats were previously acquired from Charles River
Laboratories (Montreal, PQ, Canada). All animal procedures were carried out
in accordance with the guidelines of the Canadian Council for Animal Care
(CCAC) and received ethical approval from the University of Alberta.
Animals were individually housed under controlled environmental
conditions (i.e., 22°C with alternating 12 h light and dark cycles) and received
standard rat chow and water ad libitum. Animals were randomly assigned to
one of the following four groups receiving: sham operation of the left leg only
(0-day control); 5-days, 10-days or 21-days of CLFS (n = 6, animals each
group)(See Martins, et al. 2006, 2009 for details). It should be noted that the
application of CLFS tended to elicit a compensatory effect in the contralateral
control muscles due to increased weight bearing, as previously observed
(Martins, et al., 2012; Putman, et al., 2000; Putman, et al., 2001). Therefore,
comparisons were evaluated against the 0-day animals.
2.2.2. Chronic Low Frequency Stimulation
Bipolar electrodes were implanted under general anaesthesia [75mg (kg
body wgt)-1 ketamine and 10mg (kg body wgt)-1 xylezine with 0.5mg (kg
body wgt)-1 acepromazine maleate] lateral to the common peroneal nerve of
the left hindlimb, externalized at the dorsal intrascapular region, and
connected to a small, portable stimulator. To ensure complete recovery from
surgery, a latent period of one week was allowed before CLFS was initiated.
CLFS (10Hz, impulse width 380μs, 12h day) was applied for 5-days, 10-days,
or 21-days consecutively. Strong persistent dorsiflexion was observed in the
left legs twice daily throughout the study; muscle activity was confirmed in
41
both the anterior and lateral crural compartments by palpation. 0-day
controls received a sham operation of the left leg.
The time points of stimulation for the study were chosen based on the
time course of the fast-to-slow transformation of skeletal muscle. At 5-days
mRNA and protein changes toward a slower phenotype are observed
(Jaschinski, et al., 1998; Martins, et al., 2006; Martins, et al., 2009). The final
time point of 10-days for the NFAT-isoform analysis was chosen due to the
finding of NFAT-c1 activity increases and subsequently decreases by this
time (Martins, et al., 2012). The final time point for the analysis of MyHC
transition was selected because fibre type transformations are essentially
80-90% complete after 21-days of CLFS, at both the protein and mRNA levels
(Jaschinski, et al., 1998).
2.2.3 Muscle Sampling
Upon cessation of stimulation the animals were deeply anaesthetized and the
peroneus longus muscles were excised from both hindlimbs, quickly fixed in
a slightly longitudinally stretched position and frozen in melting isopentane
(-159°C). Muscles were stored in liquid nitrogen (-196°C). Animals were
then euthanized with an overdose of Sodium Pentobarbital [Somnotol or
Euthanyl [100 mg (kg body wt)-1] (Bimedia-MTC Animal Health Inc.,
Cambridge, ON, Canada)], followed by exsanguination.
2.2.4 Fibre Type Immunohistochemistry
Frozen peroneus longus muscles were mounted in embedding medium
(Tissue-Tek O.C.T. Compound) at -20 C; 10 m-thick frozen sections were
collected from the mid belly of each peroneus longus muscle by sectioning in
the Cryostat, also at -20 C. The muscle cross-sectional area of each
individual muscle ranged from 0.5 mm2 to 2.0 mm2. Serial sections were
then immunostained for the various MyHC isoforms (i.e., BA-D5, SC-71, BF-
42
35, BF-45), Dystrophin and Laminin. Laminin sections were counterstained
with Hemotoxylin.
Immunostaining was completed according to established protocols
for MyHC isoforms (Martins, et al., 2006; Martins, et al., 2009; Martins, et al.,
2012; Putman, Jones, et al., 2003; Putman, et al., 2001). The sections were
removed from the freezer, allowed to air dry for 30 minutes, double circled
with an ImmEdge Pen (Vector Laboratories), washed once in PBS-T and
twice with PBS. Sections were then incubated for 15 min in 3% (v/v) H2O2 in
methanol and washed as before. Serial sections that were stained for BA-D5,
SC-71, BF-35, or Dystrophin were incubated for 1 h at room temperature in a
Inc.). After several washings, sections were incubated with Vectastain ABC
Reagent according to the manufacturer’s instructions (Vector Laboratories
Inc.) and reacted with 0.07% (w/v) diaminobenzidine, 0.05% (v/v) H2O2 and
0.03% (w/v) NiCl2 in 50 mM Tris-HCl (pH 7.5). All sections were
subsequently dehydrated, cleared, and mounted in Entellan (Merck,
Darmstadt, Germany).
43
2.2.5 NFAT Immunohistochemistry
Frozen peroneus longus muscles were mounted in embedding medium
(Tissue-Tek O.C.T. Compound) at -20 C; 10 m-thick frozen sections were
collected from the mid belly of each peroneus longus muscle by sectioning in
the Cryostat, also at -20 C. Immunostaining was completed according to a
modified protocol for myogenin nuclear staining (Martins, et al., 2006;
Martins, et al., 2009; Putman, et al., 2000; Putman, Jones, et al., 2003; Putman,
et al., 2001). Sections were removed from the freezer, allowed to air dry for
30 min and subsequently fixed in acetone in a whatman dish on a shaker at -
20 C for 10 min. Sections were removed from acetone, allowed to air dry
and then double circled with an ImmEdge Pen (Vector Laboratories Inc.).
Sections were washed once in PBS-T, twice with PBS and then incubated for
15 min in 3% (v/v) H2O2 in methanol, followed by another wash of PBS-T and
2 washes of PBS. Sections were then incubated for 1 h at room temperature
in blocking solution (10% goat serum, 1% BSA in PBS-T, pH 7.4) containing 4
drops of avidin-D blocking reagent (Vector Laboratories Inc.) per milliliter of
solution, followed by 2 washes of PBS. Sections were incubated overnight at
4 C with a primary antibody diluted in blocking solution and containing
biotin blocking reagent (Vector Laboratories Inc.). The next day sections
were again washed 1 time with PBS-T and 2 times with PBS, and incubated
for 1 h with biotinylated-goat-anti-rabbit-IgG (Vector Laboratories Inc.)
diluted in blocking solution. After several washings, sections were incubated
with Vectastain ABC Reagent according to the manufacturer’s instructions
(Vector Laboratories Inc.) and reacted with 0.07% (w/v) diaminobenzidine,
and 0.05% (v/v) H2O2 in 50 mM Tris-HCl (pH 7.5). All sections were
subsequently dehydrated, cleared, and mounted in Entellan (Merck,
Darmstadt, Germany).
44
2.2.6 MyHC Immunohistochemical Analysis
MyHC isoform semi-quantitative analyses was completed with a Olympus
IX70 inverted microscope (Olympus Canada Inc, Ontario, Canada) fitted with
a SPOT-RT Camera (Diagnostic instruments inc., Michigan, USA). Pictures
(see figure 3.1) were taken at 20x magnification and subsequently loaded
into a custom-designed analytical program (Putman, et al., 2000) (See figure
2.1). An area of 0.39mm2 was examined from each muscle for the various
MyHC isoforms from a representative cross-sectional area of the peroneus
longus muscles. Type-I, type-IIA, and type-IIB fibres were identified by
positive staining and type-IID(X) fibres were identified by the absence of
staining, with all antibodies directed against the various MyHC isoforms.
Figure 2.1 Example of custom designed fibre type analysis software. (Pette program)(Putman, et al., 2000). Coloured dots represent different fibre types.
45
2.2.7 NFAT Immunohistochemical Analysis
Pictures were taken of representative areas on a Leica DMRXA Compound
Light Microscope (Leica Microsystems Inc, Ontario, Canada) with a QI ClickTM
Digital Camera (Qimaging, BC, Canada) and Q Capture Pro 7 Software
& 3.11). Criteria for inclusion of positively stained myonuclei were
established by examining previous publications and only nuclei that stained
strong positive were enumerated. For all NFAT isoforms a nucleus was
considered positive when it appeared to be wholly or partially within the cell.
Further criteria differed between isoforms due to varying staining patterns
as seen in Calabria et al. (2009). NFAT-c1 and -c4 showed very strong
staining along the nuclear envelope and within the nucleus. Therefore
positive nuclei were those that had a clearly stained nuclear envelope in
addition to staining within the cell (Figures 3.6, 3.8. & 3.11). NFAT-c2 and -
c3 on the other hand displayed strong staining along the nuclear membrane,
but not within the nucleus. As such, a positive nucleus was counted when a
clearly stained membrane was identified (Figures 3.6, 3.9 & 3.10). See figure
3.5 for positive standards of myonuclear staining, which were described as a
hemotoxylin stained nucleus that was within the anti-laminin stained
basement membrane of the cell. Carl Zeiss Binocular Microscope (Carl Zeiss
Canada Ltd, Ontario, Canada) fitted with a 100x oil lens was used for total
nuclear counts as well as specific area counts. A total cross-sectional area of
1.043 mm2 was analyzed on each individual muscle in order to quantify
myonuclear counts. Numbers found for specific area counts were then
matched with MyHC analysis (Figures 3.8, 3.9, 3.10 & 3.11) using the same
custom-designed analytical program (Figure 2.1) (Putman, et al., 2000). An
area of 0.39mm2 was examined from each muscle for the various NFAT
isoforms.
46
Nuclear counts where specific area counts were performed by two
observers; an additional observer assisted with nuclear counts. Myonuclear
counts and NFAT-c2 nuclear counts were performed by observer one. NFAT-
c1 and NFAT-c4 nuclear counts were completed by observer two and NFAT-
c3 nuclear counts were completed by observer three. Specific area counts of
NFAT-c1 and NFAT-c4 were completed by observer two, while NFAT-c2 and
NFAT-c3 specific area counts were completed by observer one. Patterns of
strong, positive staining were established as the uniform standard across all
stains prior to analysis. Pilot analysis determined inter-observer reliability
was 98%.
It is important to note that the hybrid type-IIA/D(X)/B fibres were
rarely observed in the 0-day control muscles (see Figure 3.4); the frequency
of NFAT-c1, –c2, -c3 and -c4 nuclear staining at 0-day within this fibre
population was too low to conduct statistical analysis (i.e., the 4 type-
IIA/D(X)/B fibres were not positive for NFAT-c1; one of the 4 type-
IIA/D(X)/B fibres was positive for NFAT-c2; one of the 4 type-IIA/D(X)/B
fibres was positive for NFAT-c3; all of the 4 type-IIA/D(X)/B fibres were not
positive for NFAT–c4). Similarly, the pure type-IIB fibres were rarely
observed at all time points (see Figure 3.4). Thus the frequency of NFAT-c1,
–c2, -c3 and -c4 nuclear staining within pure type-IIB fibres was too low to
conduct statistical analysis. In the 0-day control muscles 1 out of 4 fibres
was positive for NFAT-c1, at 5-days 2 out of 12 fibres were positive for
NFAT-c1 and at 10-days the one pure type-IIB fibre observed did not express
NFAT-c1. In the 0-day control muscles 2 out of 4 fibres were positive for
NFAT-c2, at 5-days 9 out of 12 fibres were positive for NFAT-c2 and at 10-
days the one pure type-IIB fibre observed did not express NFAT-c2. In the 0-
day control muscles 3 out of 4 fibres was positive for NFAT-c3, at 5-days 10
out of 12 fibres were positive for NFAT-c3 and at 10-days the one pure type-
IIB fibre observed did not express NFAT-c3. In the 0-day control muscles 1
out of 4 fibres were positive for NFAT-c4, at 5-days 4 out of 12 fibres were
47
positive for NFAT-c4 and at 10-days the one pure type-IIB fibre observed did
not express NFAT-c4. Graphs representing the small number of type-IIB as
well as type-IIA/D(X)/B fibres expressing NFAT-c1, -c2, -c3 and -c4 are
shown in Figure 3.12.
2.2.8 Statistics
Data were summarized as means SEM. Differences between group means
were assessed using a one-way Analysis of Variance (ANOVA) [i.e. MyHC I x
Time (0, 5, 10 day Stimulated)]. When a significant F-ratio was found,
differences were located using the Least Significant Difference post-hoc
analysis for planned comparisons. Differences were considered significant at
P < 0.05. Data was considered a trend at P<0.10.
48
CHAPTER THREE
RESULTS
49
3.1 Animal Weights
Animals initially weighed 306±6g (mean ± SEM) and at the end of the
experiment weighed 332±6g. Weight change throughout the experiment was
similar between groups (i.e., 0-day sham operated 326±10g to 339±7g; 5-day
318±5g to 346±12g; 10-day 312±10g to 338±12g; 21-day 267±4g to
307±8g). Variations in initial weight between 21-day rats (Martins, et al.,
2006), and those studied at 0-, 5- and 10-days (Martins, et al., 2009) reflect
different initial ages of the animals, which were 3 months old (Martins, et al.,
2006) and 4 months (Martins, et al., 2009).
3.2 Fibre Type Transformation in the peroneus longus m.
The fibre type characteristics of the peroneus longus m. have not yet been
fully characterized. Armstrong and Phelps (1984) reported that the fibre
composition of peroneus longus m. of adult Sprague-Dawley rats was 48±8%
FG, 41±7% FOG and 11±2% SO. In that study they also evaluated the
oxidative capacity of fibres, using nicotinamide adenine dinucleotide
tetrazolium reductase (NADH-TR) histochemical staining, and differentiated
fibre type using myosin ATPase histochemistry. In contrast, the present
study employed more precise immunohistochemical methods and isoform-
specific MyHC antibodies to identify muscle fibre types, as shown in Figure
3.1. The proportions of FG, FOG and SO fibres reported by those authors
corresponds to type-IIB, type-IIA and type-I fibres quantified in the present
study. However, the immunohistochemical method used in the present study
allowed precise identification of type-IID/X fibres.
The proportion of pure type-IIB fibres observed in the present study
was 0.4% (Figure 3.4), which appears very low by comparison. It is,
however, important to note that myosin-ATPase histochemistry does not
distinguish between type-IIB and –IID/X fibres. Therefore, for the purpose of
comparison it seems reasonable to consider the sum of pure type-IIB (0.4%),
50
hybrid fibre types-IID/IIB (15.3%) and -IIA/IID/IIB (0.3%) (Figures 3.2 and
3.4), plus type-IID fibres (24%) (Figure 3.4). The resultant 40% of total
fibres is similar to that reported by Armstrong and Phelps (1984). The
present study extends and clarifies their earlier findings by clearly
identifying and quantifying the two fastest fibre populations within the
peroneus longus m.
In order to compare the proportion of FOG fibres reported by
Armstrong and Phelps (1984) to the corresponding fibre population
reported in the present study, it is important to apply up-to-date
nomenclature. The FOG fibre population used in that study is more
appropriately termed FO and corresponds to those fibres expressing MyHC-
IIa in the present study (Figures 3.1 & 3.2). The previously reported value of
41% (Armstrong, R. B. & Phelps, 1984) is similar to the value of 45%
observed in the present study (Figures 3.1 & 3.2). Similarly, the proportion
of pure type-I fibres quantified in the present study (i.e., 15%) (Figure 3.4)
compares very closely to the proportion of SO fibres previously reported
(Armstrong, R. B. & Phelps, 1984).
A novel finding of the present study was that 30±6% of all fibres
observed with the peroneus longus m. of 0-day controls were classified as
hybrid fibres (Figure 3.3). The co-expression of two or more MyHC-isoforms
within such a large proportion of fibres appears to be unique to the peroneus
longus m. when compared with other rat hindlimb muscles that are located
within the anterior, posterior and lateral crural compartments. Previous
studies conducted within the same laboratory that investigated mixed fast-
twitch rat tibialis anterior m. (Martins, et al., 2006; Martins, et al., 2009;
Putman, Kiricsi, et al., 2003), extensor digitorum longus m. (Putman, et al.,
2000; Putman, et al., 2001) or plantaris m. (Gallo, et al., 2008) observed a
very small proportion of hybrid fibres (1-4%) within 0-day control muscles.
51
This important difference appears to be the result of greater daily
recruitment of the peroneus longus m., as discussed in section 4.2.
CLFS induced a fast-to-slower phenotypic transition (Figures 3.1, 3.2 &
3.4). Specifically, the proportion of fibres expressing MyHC-IIa increased in
response to CLFS (P<0.002) (Figure 3.2). This was largely due to a decrease
in the proportion of IID fibres (P<0.0003) which transitioned into type-
IIA/IID hybrid fibres, as shown by progressive increases of 2.5-, 3.0- and 4.0-
fold after 5-, 10- and 21-days, respectively (P<0.03) (Figure 3.4). After 21-
days of CLFS the proportion of type-I/IIA fibres was significantly decreased,
when compared with 5- and 10-days (P<0.02); which would suggest they are
transitioning toward the Type-1 fibres (Figure 3.4). The proportion of pure
type-IIB fibres was low at 0-day and did not change in response to CLFS (i.e.,
1.2% at 5-days, 0.1% at 10-days and 0.1% at 21-days) (Figure 3.4).
CLFS-induced fibre type transitions seem to have largely occurred
after 5-days, which is earlier than previously demonstrated in the extensor
digitorum longus m. (Putman, et al., 1999, 2000) and tibialis anterior m.
(Martins, et al., 2006). Considering the large proportion of hybrid fibres that
exist within the peroneus longus m. prior to the application of CLFS (Figure
3.3), an accelerated response of this muscle to stimulation is not surprising.
Another divergent finding when compared to previous studies (Martins, et
al., 2006; Putman, et al., 1999, 2000) is the change in overall proportion of
hybrid fibre types. While there were significant changes in specific hybrid
fibre types, increases in the proportion of total hybrid fibres only
demonstrated an increasing trend (P<0.07) (Figure 3.3).
52
Figure 3.1 Representative photomicrographs of MyHC immunohistochemistry. 10 µm thick serial sections of peroneus longus m. were immunostained for MyHC-I (clone BA-D5), MyHC-IIa (clone SC-71), all MyHC’s but not MyHC-IId(x) (clone BF-35), and MyHC-IIb (clone BF-F3). Columns are as labelled: 0-day Control, 5-days of CLFS, 10-days of CLFS, and 21-days of CLFS. Scale bar represents 100 μm.
53
Figure 3.2 The proportion of muscle fibres expressing a MyHC isoform in the peroneus longus m. N = 6 at each time point. The number of fibres included in this analysis was 5,861. Data are mean ± SEM.
54
Figure 3.3 The proportion of hybrid muscle fibres simultaneously expressing two or more MyHC isoform within the peroneus longus m. N = 6 at each time point. The number of fibres included in this analysis was 5,861. Data are mean ± SEM.
55
Figure 3.4 The proportion of individual hybrid muscle fibres within the peroneus longus m. N = 6 at each time point. The number of fibres included in this analysis was 5,861. Data are mean ± SEM.
56
3.3 Myonuclear Counts of NFAT-c1, -c2, -c3, -c4
The total myonuclear content was quantified on frozen sections of the
peroneus longus m. that were immunohistochemically stained for the
basement membrane protein laminin and counterstained with Hematoxylin,
as shown in Figure 3.5. The corresponding data are summarized in Figure
3.7. An unexpected finding of this study was the approximate 2-fold greater
number of myonuclei at 0-days when compared with the tibialis anterior m.
(Martins, et al., 2006; 2009; 2012) and the extensor digitorum longus m.
(Putman, et al., 1999; 2001). Additionally, unlike the previously mentioned
studies, there were no significant time dependent changes in myonuclear
content within the peroneus longus m. in response to CLFS. Collectively, the
data show the peroneus longus m. already had a high myonuclear density
than has been observed in muscle exposed to CLFS for 20- to 50-days
(Putman, et al., 1999; 2001). This seems especially relevant to the current
data because the peroneus longus m. appears to inherently possess a
myonuclear content that exceeds a suggested threshold thought to be
necessary to support fibre type transition (Martins, et al., 2009).
As hypothesized, NFAT-c4 positive nuclei remained constant over
time. Interestingly, the absolute NFAT-c3 content (i.e., number/mm2)
increased from 0-day control to 5-day (P<0.002) and subsequently decreased
at 10-day (P<0.002). The relative NFAT-c3 content displayed the same
pattern, however did not reach significance. Since the overall myonuclear
content of the peroneus longus m. did not change (Figure 3.7) the absolute
data would be considered. The overall number of NFAT-c2 positive nuclei
showed no significant change. Whereas, consistent with previous findings
(Martins, et al., 2012) NFAT-c1 significantly increased at the 5 day time point
(P<0.0001) and then decreased at 10 days (P<0.0001) (Figure 3.7).
57
Figure 3.5 Representative micrograph of anti-laminin immunohistochemical staining, with a hemotoxylin counterstain. Myonuclei were identified as hemotoxylin positive within the circumference of the anti-laminin stain. Myonuclei are indicated by the white “”. Scale bar represents 10µm.
58
Figure 3.6 Representative micrographs of frozen sections immunohistochemically stained for the four NFAT isoforms. A) NFAT-c1, B) NFAT-c2, C) NFAT-c3, and D) NFAT-c4. Positive nuclear stains are indicated by “”. Scale bar represents 10 µm.
59
Figure 3.7 The relative (%) and absolute (number/mm2) myonuclear contents of NFAT isoforms, and total myonuclear content (number/mm2). Data are mean ± SEM.
NFATc1
NFATc2
NFATc3
NFATc4
*
c5
10
15
*
*
c
c
CLFS (days)0 5 10
total
myonuclear content
( number / mm2 )
0
200
400
600
800
1000
1200
CLFS (days)0 5 10
myonuclear content of
NFAT isoforms
( number / mm2 )
0
300
450
600
750
900
1050
*
*
c
c150
175
75
CLFS (days)0 5 10
myonuclear content of
NFAT isoforms
( % of total myonuclei )
0
30
40
50
60
70
80
90 P < 0.003 P < 0.04
P < 0.003P < 0.0001
P < 0.002P < 0.002
60
3.3 Association of Fibre type and NFAT isoform
The patterns of NFAT isoform expression were examined on serial sections
and quantified within all pure and hybrid fibre types (Figures 3.8, 3.9, 3.10
and 3.11). Criteria described in section 2.2.7 were set according to (Martins,
et al., 2012) and (Calabria, et al., 2009).
The method used to analyze the fibre type specific pattern of NFAT-c1
expression is shown in Figure 3.8 and the corresponding data are
summarized in Figures 3.12 and 3.13. The staining pattern of NFAT-c1
mirrored that of Martins et al. (2012), which was typically highly
concentrated at the nuclear envelope (Figure 3.6A). When the nuclear
localisation of NFAT-c1 was examined according to fibre types (Figure 3.12)
a clear increase was observed within type-I (P<0.02) and type-IIA/D(X)
(P<0.005) fibres at 5-days, which subsequently decreased at 10-days
(P<0.004 and P<0.003, respectively) (Figure 3.12 and Figure 3.13).
Interestingly, the type-IIA/D(X)/B fibres were rarely observed in control
muscles (0.3% of fibres, Figure 3.4).
The method used to analyze the fibre type specific pattern of NFAT-c2
expression is shown in Figure 3.9. It is important to note that the staining
pattern of NFAT-c2 (Figure 3.6B) was similar to that of NFAT-c1 but
displayed a lower staining intensity that was still clearly detectable. This
observation was consistent with the pattern shown by Calabria et al. (2009).
The corresponding data for NFAT-c2 are summarized in Figures 3.12 and
3.13. There was a significant decrease in the nuclear localization of NFAT-c2
(P<0.02) in the type-IID(X) fibres (Figure 3.13), decreasing by 4-fold from 0-
to 10-days (Figure 3.12).
The fibre type specific pattern of NFAT-c3 expression was analyzed
using the method illustrated in Figure 3.10. The intensity and pattern of
NFAT–c3 staining were similar to that of NFAT-c2 (Figure 3.6). The
61
corresponding data are summarized in Figures 3.12 and 3.14. NFAT-c3
nuclear localisation increased within type-I/IIA fibres (P<0.02) and type-
IID/B fibres (P<0.02) after 10-days of CLFS (Figure 3.12). By comparison,
NFAT-c3 nuclear localisation increased within type-IIA/D fibres after 5-days
(P<0.03) and 10-days (P<0.002) of CLFS (Figure 3.12 and 3.14).
The method used to analyze the fibre type specific pattern of NFAT-c4
expression is shown in Figure 3.11 and the corresponding data are
summarized in Figures 3.12 and 3.14. The staining pattern of NFAT-c4 was
similar to that of NFAT-c1, with a slightly higher staining intensity.
According to Calabria et al. (2009) NFAT-c4 is present in the nucleus at all
exercise states and in all fibre types, therefore the darker staining would
reflect that difference in expression. When the pattern of NFAT-c4
expression was examined according to all pure and hybrid fibre types
(Figures 3.12 & 3.14) no changes in expression were detected in response to
CLFS.
62
Figure 3.8 Representative micrographs of serial sections immunohistochemically stained to identify fibres expressing MyHC-I (BA-D5), MyHC-IIa (SC-71), all MyHC but not MyHC-IId(x) (BF-35) and MyHC-IIb (BF-F3), and to identify nuclear expression of NFAT-c1. * indicates a type-I fibre that is positive for NFAT-c1. # indicates a type-IIA fibre that is positive for NFAT-c1. Scale bar represents 10 μm.
63
Figure 3.9 Representative micrographs of serial sections immunohistochemically stained to identify fibres expressing MyHC-I (BA-D5), MyHC-IIa (SC-71), all MyHC but not MyHC-IId(x) (BF-35) and MyHC-IIb (BF-F3), and to identify nuclear expression of NFAT-c2. * indicates a hybrid type-IID(X)/IIB fibre that is positive for NFAT-c2. Scale bar represents 10 μm.
64
Figure 3.10 Representative micrographs of serial sections immunohistochemically stained to identify fibres expressing MyHC-I (BA-D5), MyHC-IIa (SC-71), all MyHC but not MyHC-IId(x) (BF-35) and MyHC-IIb (BF-F3), and to identify nuclear expression of NFAT-c3. * indicates a type-IID(X) fibre that is positive for NFAT-c3. Scale bar represents 10 μm.
65
Figure 3.11 Representative micrographs of serial sections immunohistochemically stained to identify fibres expressing MyHC-I (BA-D5), MyHC-IIa (SC-71), all MyHC but not MyHC-IId(x) (BF-35) and MyHC-IIb (BF-F3), and to identify nuclear expression of NFAT-c4. * indicates a type-IIB fibre that is positive for NFAT-c4. Scale bar represents 10 μm.
66
Figure 3.12 The proportion of individual hybrid muscle fibres within the peroneus longus m. expressing a particular NFAT isoform. N = 5 at 0-day time point, N = 5 at 5-day time point, and N = 6 at 10-day time point. The number of fibres included in this analysis was 3, 075. Data are mean ± SEM.
67
Figure 3.13 The proportion of individual hybrid muscle fibres within the peroneus longus m. expressing NFAT-c1 and -c2 isoforms. N = 5 at 0-day time point, N = 5 at 5-day time point, and N = 6 at 10-day time point. The number of fibres included in this analysis was 3, 034. Data are mean ± SEM. It is important to note that the hybrid type-IIA/D(X)/B fibres were rarely observed in the 0-day control muscles (see Figure 3.4); the frequency of NFAT-c1 and –c2 nuclear staining at 0-day within this fibre population was too low to conduct statistical analysis (i.e., the 4 type-IIA/D(X)/B fibres were not positive for NFAT-c1; One of the 4 type-IIA/D(X)/B fibres was positive for NFAT-c2). Similarly, the pure type-IIB fibres were rarely observed at all time points (see Figure 3.4). Thus the frequency of NFAT-c1 and –c2 nuclear staining within pure type-IIB fibres was too low to conduct statistical analysis. In the 0-day control muscles 1 out of 4 fibres was positive for NFAT-c1, at 5-days 2 out of 12 fibres were positive for NFAT-c1 and at 10-days the one pure type-IIB fibre observed did not express NFAT-c1. In the 0-day control muscles 2 out of 4 fibres were positive for NFAT-c2, at 5-days 9 out of 12 fibres were positive for NFAT-c2 and at 10-days the one pure type-IIB fibre observed did not express NFAT-c2.
68
Figure 3.14 The proportion of individual hybrid muscle fibres within the peroneus longus m. expressing NFAT-c3 and -c4 isoforms. N = 5 at 0-day time point, N = 5 at 5-day time point, and N = 6 at 10-day time point. The number of fibres included in this analysis was 3, 034. Data are mean ± SEM. It is important to note that the hybrid type-IIA/D(X)/B fibres were rarely observed in the 0-day control muscles (see Figure 3.4); the frequency of NFAT-c3 and –c4 nuclear staining at 0-day within this fibre population was too low to conduct statistical analysis (i.e., one of the 4 type-IIA/D(X)/B fibres was positive for NFAT-c3; all of the 4 type-IIA/D(X)/B fibres were not positive for NFAT–c4). Similarly, the pure type-IIB fibres were rarely observed at all time points (see Figure 3.4). Thus the frequency of NFAT-c3 and –c4 nuclear staining within pure type-IIB fibres was too low to conduct statistical analysis. In the 0-day control muscles 3 out of 4 fibres was positive for NFAT-c3, at 5-days 10 out of 12 fibres were positive for NFAT-c3 and at 10-days the one pure type-IIB fibre observed did not express NFAT-c3. In the 0-day control muscles 1 out of 4 fibres were positive for NFAT-c4, at 5-days 4 out of 12 fibres were positive for NFAT-c4 and at 10-days the one pure type-IIB fibre observed did not express NFAT-c4.
69
CHAPTER FOUR
DISCUSSION
70
4.1 Introduction
The purpose of the present study was to investigate the relationship between
the activity-induced transition of individual fibre types, and the preceding
expression patterns of various NFAT-isoforms within mammalian skeletal
muscle (i.e., NFAT-c1, NFAT-c2, NFAT-c3 and NFAT-c4). Previous studies
have reported correlations between muscle fibre types and the pattern of
NFAT-isoform expression in normal control muscles (Calabria, et al., 2009;
Meissner, Umeda, et al., 2007; Mutungi, 2007). However, the question
remains as to whether muscle training can induce predictable changes in the
patterns of NFAT-isoform expression, and associated nuclear translocation,
which corresponds to the newly transformed muscle fibre phenotype.
A unique characteristic of this study is the comprehensive
characterization of the immunohistochemically identified fibre type
distribution of the peroneus longus m. at rest and with stimulation. Contrary
to the current hypothesis, the proportion of fibres expressing each NFAT
isoform at rest did not differ between fibre types. When stimulated with
CLFS an overall shift toward a slower phenotype was observed. The
proportion of fibres expressing NFAT-c1 and NFAT-c4 with CLFS happened
as expected. Considerable transitions occurred within the fibre populations
expressing MyHC-IIa and –IId(x). A corresponding increase in NFAT-c3 and
decrease in NFAT-c2 nuclear localisation, respectively, suggests a role for
NFAT-c2 in the maintenance of the IID(X) phenotype and a role for NFAT-c3
in the transition toward type-IIA fibres. Overall, changes in NFAT-c2 and -c3
were consistent with having important roles in the transformation and
maintenance of the more oxidative fast fibre populations.
4.2 The Unique Composition of the Peroneus Longus m. at
Rest
71
The rat peroneus longus m. is an infrequently studied rat mixed fast-twitch
muscle in training adaptation research and as such, the fibre type
composition has not been fully elucidated. Previous research has reported
the composition to be 48±8% FG fibres (i.e., type-IIB, type-IID(X)), 41±7%
FOG fibres (i.e., type-IIA, type-IID(X)), and 11±2% SO (i.e., type-I)
(Armstrong, R. B. & Phelps, 1984). In contrast, the current investigation has
found 39±7% FG fibres, 46±3% FOG fibres, and 15±4% SO fibres. As
discussed in section 1.3, the discovery of MyHC-IId(x) (Bar & Pette, 1988)
and the corresponding type-IID(X) fibres (Schiaffino, et al., 1989) occurred
after the study of Armstrong and Phelps (1984). Therefore, the more precise
immunohistochemical methods could partially account for the difference in
proportions. Similar to current numbers, the proportion of type-I fibres in
the peroneus longus m. in cat was reported to be 11.8 - 14.6% (Kernell &
Hensbergen, 1998). In sheep, the peroneus longus m. is considered a
locomotory muscle (Watanabe & Suzuki, 1999), if the same is true of rat it
could explain the resting fibre type proportions. Kernell & Hensbergen
(1998) measured daily muscle activity of cat muscles and found the peroneus
longus m. was in use ≤9.5% of the day whilst, the extensor digitorum longus
m. was only in use ≤1.9% of the day. Hence, the fibre type composition of the
peroneus longus m. at rest is substantially different than the more commonly
studied tibialis anterior m. and extensor digitorum longus m. The proportion
of type-IIA fibres is substantially higher in the peroneus longus m., while the
proportion of type-IIB fibres is considerably lower. While the peroneus
longus m. is described as a mixed fast-twitch muscle, but its composition is
considerably slower than other mixed fast twitch muscles. Previous research
has directly compared the extensor digitorum longus m. with the peroneus
longus m. and found similar results. The area occupied by the type-IIB fibres
was less than 5% in the peroneus longus m, whereas it was approximately
70% in the extensor digitorum longus m. (Ranatunga & Thomas, 1990).
72
A novel finding of the present study was the large proportion of
hybrid fibres in control peroneus longus m. which was 30±6%. In contrast,
Martins et al. (2009) found the proportion of hybrid fibres in control tibialis
anterior m. was only 4%. Hybrid fibres are an indication of fibre type
transition and are normally seen to increase in a time dependent manner
with stimulation. This large discrepancy in hybrids between the tibialis
anterior m. and the peroneus longus m. would suggest transition is happening
at rest and being a locomotory muscle, transition at rest would not be totally
unexpected. This constant state of plasticity would allow for a muscle that is
more readily adaptive to stimulation.
There are many inherent differences between slow- and fast-twitch
muscles at rest as discussed earlier. Major variations are present in Ca2+
concentration kinetics, including a higher resting cytosolic free [Ca2+] in
slower, than faster muscles (Carroll, S. L., et al., 1997; Carroll, S. L., et al.,
1999; Fraysse, et al., 2003; Fraysse, et al., 2006); a larger sarcoplasmic
reticulum in fast muscles (Trinh & Lamb, 2006); and a higher peak of Ca2+
transients and quicker rate of decline in fast muscles (Baylor & Hollingworth,
2003). Given the different traits in Ca2+ kinetics at rest and the mixed
composition of the peroneus longus m., early phenotypic transitions observed
in the present study were not surprising.
4.3 CLFS INDUCED TRANSFORMATION OF THE PERONEUS LONGUS M.
CLFS is a well-characterized model of fast-to-slow fibre type transformation
in rat that creates sustained low-amplitude increases in [Ca2+] known to
preferentially activate the Cn-NFAT pathway (Dolmetsch, et al., 1997;
Timmerman, et al., 1996). Previously, our lab has analyzed the changes in
the tibialis anterior m. with stimulation (Martins, et al., 2006; 2009; 2012).
At 21-days of stimulation a significant increase in MyHC-I and a concurrent
decrease in MyHC-IIb and MyHC-IId(x) were observed (Martins, et al., 2006),
73
with considerable fibre transformation occurring at the 10-day time point
(Martins, et al., 2009). Comparatively, in the peroneus longus m. substantial
fibre type adaptation is noticed at the 5-day time point. The rapid
transformation of the peroneus longus m. is most likely due to the unique
plastic nature of this particular muscle. The large number (~30%) of hybrid
fibres at rest indicates transformation was already occurring without any
external stimuli, therefore it is not surprising that it adapted more readily to
CLFS. Specific findings show a significant increase in fibres expressing
MyHC-IIa throughout the time course (Figure 3.2), which is in agreement
with Martins et al. (2009). An increase in type-IIA/D(X) fibres, along with a
decrease in type-IID(X), and type-IID(X)/B fibres, indicates fast-to-slow fibre
type transformation occurred.
4.4 NFAT-c2 and NFAT-c4 are Important for Cell Survival
and Fibre Type Maintenance
NFAT-c2 has been found active in many tissues throughout the body in the
regulation of cell size and/or number (Horsley & Pavlath, 2002; Ikeda, et al.,
2006; Perotti, et al., 2012). In skeletal muscle Calabria et al. (2009)
suggested that at rest NFAT-c2 nuclear localisation is imperative for the
expression of MyHC-I, -IIa, and -IId(x), and with electrical stimulation NFAT-
c2 is required for transition of the fastest type-IIB fibres into slower type-
IID(X) and eventually type-IIA fibres. Findings of this investigation suggest a
contrary role for NFAT-c2 with stimulation. The overall decrease in this
particular isoform throughout the time course would seem to indicate a
larger role for NFAT-c2 at rest than during transformation. Specifically, the
proportion of NFAT-c2 positive nuclei significantly drops at 10-days in type-
IID(X) fibres, this change coincides with a decrease in type-IID(X) fibres and
a subsequent shift to a slower phenotype. NFAT-c2 plays a significant role
during growth in the regulation of myoblast to myotube cell fusion (Horsley,
et al., 2001; Horsley, et al., 2003; Horsley & Pavlath, 2003), which leads to
74
growth and maintenance of the muscle fibre. A similar function for NFAT-c2
could extend to adult skeletal muscle, suggesting NFAT-c2 has a larger role in
fibre maintenance and cell survival, rather than transformation without
injury.
In skeletal muscle, Calabria et al. (2009) reported NFAT-c4 to be
required for the maintenance of all fibre types during rest and is involved in
the suggested default type-IIB/IID(X) fibre type (Butler-Browne, et al., 1982).
Results of the present study support that proposition, with NFAT-c4 levels
staying relatively consistent during stimulation. In other tissues, NFAT-c4
has been implicated in cell apoptosis (Li, L., et al., 2013), and may underlie
activity-dependent neuronal plasticity throughout the adult brain (Bradley,
et al., 2005). NFAT-c4 has also been shown to be involved in cell survival
(Vashishta, et al., 2009), which could be the case in skeletal muscle as well.
Overall, current findings as well as findings in other tissues support the
notion that NFAT-c4 is involved in the maintenance of a “base fibre type.”
4.5 NFAT-c1 and NFAT-c3 are Regulators of fast-to-slow
Fibre Type Transition
NFAT-c1 is the most widely studied NFAT isoform in adult skeletal
muscle to date. Previous research has focused on NFAT-c1 as a key regulator
of the slow-gene program in adult muscle at rest (McCullagh, et al., 2004;
Mutungi, 2008; Shen, et al., 2006), and with stimulation (Calabria, et al.,
2009; Martins, et al., 2012; McCullagh, et al., 2004; Tothova, et al., 2006). The
present study used Chronic Low-Frequency Stimulation, which induces the
slow-gene program (Martins, et al., 2012; Pette, 1991; Putman, et al., 1999,
2000) to investigate NFAT-c1 activity. The pattern of stimulation is
important in the activation and dephosphorylation of NFAT-c1 (Liu, J., et al.,
2001; Martins, et al., 2012; McCullagh, et al., 2004; Tothova, et al., 2006). As
such, CLFS provides a slow, consistent increase in Ca2+ transients that have
75
been shown to activate the Cn-NFAT signalling pathway (Chin, et al., 1998;
Tothova, et al., 2006).
Findings of the present study indicate a significant role for NFAT-c1 in
the early transition from a faster-to-slower fibre type, evidenced by the
overall increase in NFAT-c1 at 5-days of stimulation and subsequent
decrease at 10-days. In particular, there is a substantial increase in NFAT-c1
nuclear localisation at 5-days in type-I and type-IIA/D(X) fibres. The
subsequent increase in type-IIA/D(X) fibres along with a concurrent
decrease in type-IID(X) and type-IID/B fibres, indicates fast-to-slower fibre
type transformations occurred. These findings are in agreement with
Martins et al. (2012), who showed a significant increase in NFAT-c1 nuclear
localisation within the tibialis anterior m. after 5-days of CLFS and a decrease
after 10-days.
In addition, Martins et al. (2012) demonstrated the hindrance of
NFAT-c1 activity with the blockage Nitric Oxide and downstream GSK-3
dephosphorylation. This finding would implicate GSK-3 as a main export
kinase of NFAT-c1, and in turn a regulator of NFAT-c1 activity. Confirming
the postulation that differential regulation of the four calcium activated
NFAT isoforms could be largely due to the different export and maintenance
kinases, rather than activators. In T-cells, the domain of NFAT-c1 that is
targeted by calcineurin is phosphorylated and inactivated by the c-Jun NH2-
terminal kinase (JNK) (Chow, et al., 2000). Measuring levels of the other
NFAT isoforms with Nitric Oxide blockade would provide insight into
whether isoforms other that NFAT-c1 are exported by GSK-3 , another
kinase, or a combination. Martins et al. (2012) and Calabria et al. (2009)
both implicated NFAT-c1 in the final transition from type-IIA to type-I. The
current data cannot undisputedly discern the function of each NFAT isoform
independent of the others, however the significant amount of NFAT-c1 at 5
days in type-I fibres suggests it is the major isoform regulating MyHC-I.
76
NFAT-c3 seems to function somewhat differently than NFAT-c1.
Calabria et al. (2009) used a model based on NFAT-isoform specific siRNAs
and concluded that NFAT-c3 was important for the maintenance of MyHC-I,
MyHC-IIa, and MyHC-IId(x), with a stronger influence on MyHC-IIa than
MyHC-IId(x). With stimulation, NFAT-c3 translocated into the nucleus in
response to slow and fast patterns of stimulation (Calabria, et al., 2009). In
this investigation, NFAT-c3 increased in hybrid type-I/IIA, type-IIA/D(X) and
type-IID(X)/B fibres. Consistent with the hypothesis tested in this study the
nuclear localisation of NFAT-c3 occurred within type-I/IIA and type-IIA/D(X)
fibres, which are both transforming hybrid fibres expressing MyHC-IIa.
Furthermore, NFAT-c3 appeared in the nuclei of type-IID(X)/B fibres, which
are known to transform into type-IIA/D(X)/B fibres and then to type-
IIA/D(X) fibres, which are also MyHC-IIa expressing hybrid fibre types
(Figures 3.12 & 3.14).
Dephosphorylation and nuclear localization of NFAT-c3 appear to be
activity dependent. Shen et al. (2006) found, in cultured CD-1 mice flexor
digitorum brevis fibres, that NFAT-c3 did not shuttle in and out of the nucleus
at rest even with the blockage of CRM1 (an export kinase) and suggested that
stimulation is required for the nuclear localisation of NFAT-c3 (Shen, et al.,
2006). In neuronal cells, NFAT-c3 rapidly dephosphorylated and
translocated to the nucleus in response to Ca2+ transients (Ulrich, et al.,
2012). Additionally, the present study found that NFAT-c3 is activity
dependent due to the increase in nuclear localization in response to 10-days
of CLFS. As for a role in resting fibre type maintenance, the proportion of
NFAT-c3 positive nuclei is significant in all fibres not expressing MyHC-IIb
when compared to NFAT-c1 levels. Overall, the results of this study indicate
that NFAT-c3 is involved in the promotion of type-IIA fibres and is important
for CLFS induced fibre type transitions.
77
4.6 CONCLUSIONS
The objective of this study was to better understand the mechanisms
involved in skeletal muscle adaptation to environmental stimuli as well as
the pathways that control fibre type transitions. This hypothesis driven basic
study was designed to further characterize the cellular signals regulating
activity-induced fibre type transitions. Results suggest that all four NFAT
isoforms are involved in skeletal muscle in non-redundant roles. In contrast
to the hypothesis, the proportion of fibres expressing a specific NFAT isoform
at rest did not differ between fibre types. In response to CLFS, however,
NFAT-c1 and NFAT-c3 were involved in the fast-to-slow fibre type
transformation in peroneus longus m. Specifically, NFAT-c3 was involved in
transformation surrounding type-IIA fibres. Concurrent data were
consistent with NFAT-c2 being involved in determination of the type-IID(X)
phenotype at rest. As expected, NFAT-c4 remained stable with stimulation,
consistent with a role is in cell maintenance. Thus, CLFS-induced shifts in the
nuclear localization of NFAT-isoforms within transforming muscle fibres
provide new insights regarding the specific MyHC-isoform likely targeted.
Future investigations should include the earlier and later time points
of CLFS. Since the fibre type transition in this analysis occurred more readily
than some of the other mixed fast-twitch muscles, earlier time points (i.e., 1-
and 2-days of CLFS) would elucidate the preceding changes in NFAT. The
later time point (i.e., 21-days of CLFS) would provide insight into NFAT at
rest in a phenotypically slower muscle. As well, valuable information could
be gained through additional quantitative assays that would provide further
insight as to the location of the 4 NFAT isoforms at rest and in response to
stimulation (e.g., Phosphorylation status of each NFAT isoform, and single
fibre analysis). Lastly, further investigation into the many export and
maintenance kinases that act on NFAT would provide insight into the overall
balance of NFAT content in the nucleus.
78
REFERENCES
Abdelmalki, A., Fimbel, S., Mayet-Sornay, M. H., Sempore, B., & Favier, R.
(1996). Aerobic capacity and skeletal muscle properties of normoxic
and hypoxic rats in response to training. Pflugers Arch, 431(5), 671-
679.
Al-Shanti, N., & Stewart, C. E. (2009). Ca2+/calmodulin-dependent
transcriptional pathways: potential mediators of skeletal muscle
growth and development. Biol Rev Camb Philos Soc, 84(4), 637-652.
Allen, D. L., & Leinwand, L. A. (2002). Intracellular Calcium and Myosin
Isoform Transitions. The Journal of Biological Chemistry, 277(47), 7.
Allen, D. L., Sartorius, C. A., Sycuro, L. K., & Leinwand, L. A. (2001). Different
Pathways Regulate Expression of the Skeletal Myosin Heavy Chain
Genes. The Journal of Biological Chemistry, 276(47), 9.
Andersen, J. L., Klitgaard, H., & Saltin, B. (1994). Myosin heavy chain isoforms
in single fibres from m. vastus lateralis of sprinters: influence of
training. Acta Physiol Scand, 151(2), 135-142.
Armstrong, R. B., & Phelps, R. O. (1984). Muscle fiber type composition of the
rat hindlimb. Am J Anat, 171(3), 259-272.
Armstrong, S. P., Caunt, C. J., Fowkes, R. C., Tsaneva-Atanasova, K., & McArdle,
C. A. (2009). Pulsatile and sustained gonadotropin-releasing hormone
(GnRH) receptor signaling: does the Ca2+/NFAT signaling pathway